MuseBench (Music Score and Performance Understanding)
Models answer music theory questions, read piano scores, and interpret recorded performances. The authors report accuracy separately for text, images, and audio.
Microsoft Research published the MuseAgent technical report. It compares models across 28 task types and reports 86.7% text accuracy for GPT-4.1.
- Made by
- Qihao Zhao and colleagues
- Kind
- Audio understanding and voice
- Level
- University
- Scored
- Checked automatically
- Answers
- Multiple choice
- Measures
- Capability
- Subject
- Arts
- Language
- English
- Interaction
- Single turn
- Refresh
- One fixed set
- Run by
- A host
- In views
- Vision, Audio and speech
- Last model added
- March 2025
- Scores refreshed
- November 21st, 2026
GPT-4o (snapshot unspecified) tops the leaderboard with 68.5% Qwen2.5-Omni-7B leads with 52.8%
Best score per entry from the benchmark's paper, read on November 21st, 2026. A paper’s table doesn’t add newer models. The paper reports no overall score, so this is the plain mean of text, audio, and image accuracy. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. The best model we track is Qwen2.5-Omni-7B, at 52.8%. This view shows only models we track, each at its best.
- 1 GPT-4o (snapshot unspecified) 68.5%
- 2 Gemini2.5-Pro (snapshot unspecified) 66.4%
- 3
Qwen2.5-Omni-7B 52.8%
Source: The MuseBench (Music Score and Performance Understanding) paper
- 1
Qwen2.5-Omni-7B 52.8%
Source: The MuseBench (Music Score and Performance Understanding) paper
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | GPT-4o (snapshot unspecified) | 68.5% | Official leaderboard | ||
| 2 | Gemini2.5-Pro (snapshot unspecified) | 66.4% | Official leaderboard | ||
| 3 | | 52.8% | Mar 2025 | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 52.8% | Mar 2025 | Official leaderboard |
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official MuseBench (Music Score and Performance Understanding) leaderboard
No lab model card reports it yet
We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.
More in Audio and video
- Video-MMMU In use 300 expert-level lecture videos with questions testing perception, comprehension and adaptation of knowledge. Accuracy.
Qwen 3.8 Max 88.7%
- LVBench In use Hours-long video understanding with multiple-choice questions on reasoning, grounding and summarization.
Gemini 4 Argon 91.7%
- MMVU In use 3,000 expert-annotated questions on videos across 27 disciplines. Multiple-choice and open-ended accuracy.
Qwen 3.8 Max 82.4%
- MLVU In use Multi-task long video understanding (minutes to hours) with multiple-choice and generation tasks. Accuracy.
Qwen 3.8 Max 90.8%
- FLEURS In use Models transcribe read speech in 102 languages built from FLoRes-101 sentences. Model cards usually report word error rate, so lower scores are better.
Phi-4-multimodal 4.00%
- VoiceBench In use Spoken instructions with speaker, environment and content variation. Voice assistant responses scored by GPT judge or rules.
Qwen3.5-Omni-Plus 93.1%
Data audit
7 of 9 checks pass, 2 warnings. Look first at the Task access check. We don’t know whether the tasks are public. Every benchmark
Catalog record 1 warning 1 pass
- Warning Task access We don’t know whether the tasks are public.
- Pass Confidence Checked against a primary source.
Scores 1 warning 2 pass 2 notes
- Pass Independent runs Scores from the benchmark’s paper (3 entries).
- Note Models scored Only 1 model we track, plus 2 other entries on its leaderboard.
- Warning Newest score The newest came 20 months ago, in March 2025, for Qwen2.5-Omni-7B.
- Note Frontier models None of its models are on the capabilities index.
- Pass Headroom The best is 52.8%, under 85%, so it still separates models.
Leaderboard import 3 pass
- Pass Importer We read the board on arxiv.org: 1 row for models we track and 2 other entries at the last read.
- Pass Last read November 21st, 2026.
- Pass Read worked The last read worked.
Consistency 1 pass 1 note
- Note Lab claims No model has both a lab number and an independent run.
- Pass Known issues No published audit found problems.